Keystroke motion identification method based on cosine similarity

Abstract

The invention provides a keystroke motion identification method based on cosine similarity, which relates to the field of network safety based on negative selection algorithm. Based on the keystroke dynamics, the biological immunization process can be simulated by means of the negative selection algorithm. Through the abstractness of immune cells, non-self detection can be completed by the detector, and the aim of distinguishing truth from falsehood can be achieved. The invention is advantageous in that the keystroke dynamics is applied, and with multiple attributes, the bottleneck of the negative selection algorithm in a high-dimensional space can be broken through; the high-dimensional negative selection algorithm has great advantage in the self recognition and non-self recognition in the complicated compound application scene.

Description

technical field [0001] The invention relates to the field of network security based on a negative selection algorithm, in particular to a keystroke recognition method based on cosine similarity. Background technique [0002] It is clear that digital identities represent an important advancement in our society. However, the dissemination of these identities has resulted in increased data exposure and identity theft. , identity theft occurs when a person uses another person's personal information to illegally pretend to be that person. Another promising approach for curbing this problem is to use behavioral intrusion detection systems, which detect anomalous behavior as potential intrusions. [0003] Among the aspects in which a user may be analyzed, keystroke dynamics are studied. This work demonstrates that an immune negative selection algorithm (NSAS) is applied to identify users based on their typing rhythm. These algorithms are novelty detectors that use only positive...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the current situation that the security of digital identity is not high but the application range is extremely wide, the present invention provides a cosine similarity-based method that uses negative selection algorithm to simulate the biological immune process and abstracts immune cells into detectors to complete non-self detection. keystroke recognition method

Images